Changing the log levels in error scenarios

[vidyadharamurthy]

No description provided.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (main@740dd14). Learn more about missing BASE report.
⚠️ Report is 548 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #6832   +/-   ##
=======================================
  Coverage        ?   90.01%           
  Complexity      ?     7090           
=======================================
  Files           ?      803           
  Lines           ?    21444           
  Branches        ?     2092           
=======================================
  Hits            ?    19302           
  Misses          ?     1477           
  Partials        ?      665           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[vidyadharamurthy]

@jkwatson - Request your review

[jack-berg]

Given that this PR only changes the log level, is this test trying to add missing coverage for existing behavior? Or verify logs are emitted at the intended level?

If the latter, I suggest we skip this. We do have test tooling to verify log messages (for example), but asserting the log level seems rather specific.

[trask]

wdyt of using ApiUsageLogger instead? (though I'm ok with this PR since it only affects opentracing-shim)

[jack-berg]

wdyt of using ApiUsageLogger instead? (though I'm ok with this PR since it only affects opentracing-shim)

ApiUsageLogger logs messages with an AssertionError, which would be pretty noisy and hard miss, especially if logged at the warning level as proposed in this PR.

I'm not opposed to it since I think that anytime one of these things occurs it indicates that some instrumentation is misusing the API, but just wanted to call it out.

[trask]

ApiUsageLogger logs messages with an AssertionError, which would be pretty noisy and hard miss, especially if logged at the warning level as proposed in this PR.

ah, I didn't even see that ApiUsageLogger takes an optional level, it doesn't look like we ever use that method and only ever use the default (FINEST)

[otelbot]

Thank you for your contribution @vidyadharamurthy! 🎉 We would like to hear from you about your experience contributing to OpenTelemetry by taking a few minutes to fill out this survey.